In general, the goal of anti-cancer immunotherapy has been to identify stable antigens that are highly expressed but not shed or secreted from tumor cells, which antigens can then be used as the basis of immunotherapy, e.g., as the antigen in a cancer vaccine or as a target for antibody-based cancer therapy. Optimally, such tumor antigens would also be ones that elicit an immune response that is acceptably specific for the cancerous target cells, so as to reduce deleterious side effects that can result from cross-reactivity with non-cancerous cells of the subject being treated. Where cross-reactivity affects cells that can be repopulated, it may be acceptable to relax this requirement for the specificity of immunotherapy.
For example, although other antibodies are available for use in treating leukemias/lymphomas, the current standard for monoclonal antibody (mAb) therapy of non-Hodgkins lymphoma is RITUXIMAB™, a chimeric murine/human mAb that recognizes CD20 antigen. CD20 is highly expressed in most mature B cells and B-cell lymphomas, exhibits relatively slow modulation of expression or antigenic determinants, and is not shed or secreted. Although this antibody also binds CD20 on non-cancerous B cells, this cell population can be restored, e.g., through supportive treatment with immune enhancing therapeutics such as granulocyte-macrophage colony-stimulating factor (GM-CSF), erythropoietin (EPO), etc.
Fc regions of antibodies binding to cell surface antigens can mediate complement deposition and cell lysis (complement dependent cytotoxicity or CDC) or antibody-dependent cellular cytotoxicty (ADCC) by activating natural killer (NK) cells. Although not as common, antibodies can also be cytotoxic by binding to cell surface antigens that affect a signaling pathway leading to apoptosis. For example, although the mechanism of action of RITUXIMAB™ is not completely understood, it appears to exert its cytotoxic effects on CD20-positive tumor cells by a combination of antibody-dependent CDC, ADCC, and by activating cellular signaling pathways that lead to apoptosis. With the success of Rituximab, several other cellular antigens have been targeted with mAbs including CD22, CD30, and CD80.
In addition to passive immunotherapy through antibody administration, several active immunization strategies have also been explored. Exemplary cancer vaccines involve administration of a tumor antigen so as to elicit humoral antibody and/or cellular immune responses that are able to activate complement, and opsonophagocytotic killing of tumor cells. Exemplary vaccine compositions include those based on tumor cell lysates, and tumor-specific antigens (e.g., proteins, gangliosides (Tai, T., et al. Int J Cancer, 1985. 35:607-12), anti-idiotype immunoglobulin (Ig), etc. (Foon et a 1. J Clin Oncol, 2000. 18: 376-84) or peptide fragments of tumor-specific or overexpressed proteins that are derived from non-autologous and autologous cancers of the same type (Morioka et al. J Immunol, 1994. 153: 5650-8; Morioka, N., et al., Mol Immunol, 1995. 32: p. 573-81).
One limitation of these approaches has been that, although the target antigens are typically more highly expressed in tumor cells, they are nonetheless autoantigens and, thus, poorly immunogenic. Thus, cancer vaccines often employ various strategies for enhancing immunogenicity of the cancer antigen, e.g., combination with adjuvants, administration with a cytokine(s), linkage to to carrier proteins, and use in pulse-activation of mature dendritic cells in vitro. A more recent trend has been to develop more elaborate vaccination strategies that are tailored to the patient in which autologous dendritic cells are isolated, stimulated with tumor lysates or peptides and reinjected either alone or in combination with potent immunostimulatory cytokines (e.g., GM-CSF, interleukins IL-2 and IL-12, interferon gamma).